Research On Energy Management Of Household Microgrid Considering The Prediction Error Of Photovolotic Power

With the increasing household electricity consumption,the gradual introduction of distributed energy and the increasing popularity of building automation equipment and two-way communication infrastructure,the research on household energy management has attracted much attention in recent years.The research object of this paper is the user side microgrid including photovoltaic,energy storage and typical household electrical equipment.Considering photovoltaic power,on the basis of the prediction error,give full consideration to the family operation of electrical equipment,the distributed generation system,the energy storage equipment and electric vehicle charging and discharging situation,design the joint optimization of scheduling strategy of energy storage and electric vehicles,to establish the optimization goal of the user side and grid side of give attention to two or morethings,to reduce the electricity cost,improving the structure of the system can use and improve the reliability and stability of power grid,the purpose of the main research content is as follows:(1)The basic model of household micro-grid power system is established,including the structure model of micro-grid and the working model of typical power equipment.Among them,typical electrical equipment takes HVAC system,electric water heater and electric vehicle as the research object.(2)In order to improve the accuracy of photovoltaic power prediction and enhance the reliability of the optimal decision scheme of energy management system,a super short term prediction method of photovoltaic power generation based on long and short term memory network was proposed.A multivariate time series prediction model based on LSTM network was established,and Pearson correlation coefficient was used to analyze the correlation between photovoltaic power and meteorological factors,so as to find the core factors affecting photovoltaic power output,determine the input variables of the prediction model,and realize the prediction of photovoltaic power.Aiming at the error of prediction results,the gaussian mixture model is used to analyze the prediction errors of different distribution characteristics.(3)In order to give full play to the idle capacity of ev batteries,improve the energy utilization rate,and integrate ev into the energy management interaction,a joint optimal scheduling strategy of energy storage and ev is proposed.Meanwhile,charging and discharging management of ev and user’s travel demand are taken into account.On the basis of considering the prediction error of photovoltaic power,a micro-grid multi-objective optimization model was established with the optimization objectives of minimum power consumption cost and minimum load fluctuation.To fully explore the potential of energy regulation of electric vehicles.Set up two kinds of optimization strategy,and on the MATLAB platform by using the NSGA-Ⅱoptimization algorithm for the proposed joint optimization scheduling strategy is verified.